GA applications to RSM based models for burr size reduction in drilling

Abstract

Application of Genetic Algorithm (GA) for reduction of burr size in drilling process, models using Response Surface Methodology (RSM) were presented. Since deburring processes are not yet well automated, understanding of the burr formation in drilling and its dominant parameters is essential for controlling the burr size at the production stage itself. Second order mathematical models of burr height and burr thickness are developed using central composite rotatable design of experiments for drilling of mild steel work pieces. The effect of cutting speed, feed, drill diam, point angle and clearance angle on burr height and thickness has been investigated. The developed RSM models are then employed with GA, which is a search algorithm based on natural selection, to minimize the burr size. The advantages of GA to obtain optimum process parameters in a multi-objective non-linear problem by proper design of fitness function involving burr size are demonstrated through simulation results.